Skip to content
LIVE
The Executives BriefThe Executives Brief

Climate Models Assume 92% Carbon Capture Success. Only 12 Projects Have Ever Been Built.

A ProPublica and Drilled investigation finds the world's three most influential climate models baked in unproven carbon capture assumptions, turning net-zero pledges into unpriced risk for boards and lenders.

ByLama Al-RashidTechnology Correspondent, The Executives Brief
·5 min read
Climate Models Assume 92% Carbon Capture Success. Only 12 Projects Have Ever Been Built.
Executive summary

A ProPublica and Drilled investigation, drawing on more than 100 experts and the leaders of the Dutch, German and Austrian climate models, found those models assume up to a 92% carbon capture success rate even though only 12 large-scale storage projects have ever been built and at least seven underperformed or failed. For executives, that gap turns carbon-neutral labels, offsets and CCS-heavy transition plans into disclosure and capital-allocation risk that boards can no longer treat as settled science.

The number at the center of global climate policy is 92%. That is the carbon capture success rate the world's most influential climate models assume, according to a ProPublica and Drilled investigation. The number of large-scale geologic storage projects ever built, after three decades of development, is 12. At least seven of those underperformed or failed. Last year, the world buried less CO2 underground than a single large power plant emits annually.

Those two facts sit on a collision course, and the crash site is every "carbon neutral," "net zero" and "climate pledge friendly" label you have seen on airline booking screens, shipping boxes, electronics and grocery aisles. The premise behind those labels is that pollution from continued fossil fuel burning can be canceled out by capturing carbon and locking it away. CCS aims to trap CO2 that would otherwise warm the atmosphere and pump it thousands of feet below Earth's surface, where, ideally, it stays put. The oil industry has long favored the idea, and at the industry's prodding the United Nations has effectively endorsed the notion that the technology can work at massive scale. At the UN's annual flagship climate summit in Antalya, Turkey, a central foundation of the negotiations is the expectation that we can bury billions of tons of CO2 a year, an operation that would rival today's entire global oil industry within just a couple of decades.

The reporter spoke to more than 100 experts, including several members of the group that builds and runs the models. Their simulations map how choices like retiring coal plants or electrifying vehicles could prevent warming catastrophic enough to submerge cities like Miami. Which means if they are wrong, Miami could pay the price. One chart laid bare the gap: it compares a prominent model's projections for CCS with its projections for solar power, a technology so abundantly tested that you can likely spot a panel out the nearest window. Solar electricity production has grown rapidly, far outpacing projections. Carbon capture capacity has never come close. The modeled pathways come from text and tables in the International Energy Agency's Energy Technology Perspectives and World Energy Outlook reports, and from correspondence with the IEA.

Why the disparity? The models are opaque. Unlike most scientists, who trade in public, peer-reviewed studies, modelers are not required to document every assumption they build into their influential simulations. The reporter tracked down the leaders of the world's three most influential climate models, based in the Netherlands, Germany and Austria. The findings: models assume it is easy and efficient to capture carbon. It is not. Public data is hard to come by, but several analyses have found real capture rates far lower than the industry has promised. Yet the models calculate as much as a 92% success rate. Keywan Riahi, who leads the Austrian model, insisted that capturing concentrated CO2 from industrial emissions "is not high-tech, this is very easy to do." Riahi said the models assume a high capture rate because they factor in incentives such as payouts from carbon taxes and large government subsidies.

Real-world practice says otherwise. In project after project, operators have struggled to inject as much carbon into the earth as they say they can. At Chevron's Gorgon natural gas processing plant in Australia, billed as the largest CCS operation on Earth, clogged wells and pressure spikes have forced operators to release most of their CO2 straight into the air. Only the German model, led by Jessica Strefler, assumed any limit to how much CO2 operators could inject underground at once. But she told the reporter there is no good data to work from, and she pegged the limit based on "expert interviews" with people in the industry.

Then there is the assumption that the Earth is a bottomless geologic landfill. Modelers took that cue from a rosy 2005 UN report, written at the behest of Saudi Arabia and allied petrostates, that portrayed underground storage space as practically limitless. The UN's climate science body has said it stands by the report, and scientists who contributed defended their work. Authors and editors from Saudi Aramco, Chevron and Shell joined scientists in writing it; the companies did not respond to requests for comment. Just last year, a paper in the journal Nature warned that the earth's capacity to absorb CCS could be far more limited once you rule out pumping the gas beneath cities, into earthquake zones and under the deepest parts of the oceans. If CCS ever worked at the scale modelers imagine, the paper found, the Earth could run out of room as soon as early next century. Riahi, a co-author on that paper, conceded that although his model presumes the earth's capacity for carbon storage is literally infinite, the real potential may be "smaller."

Leaks are another blind spot. Scientists warn that even tiny annual leak rates over long time horizons could undermine a storage site's climate benefit. The investigation found two large-scale sites that have leaked, one all the way to the surface. Asked whether his model factors in that risk, Detlef van Vuuren, leader of the Dutch model, gave a one-word answer: "Nope." The models also assume CCS gets more affordable over time, though real-world costs have steadily climbed. Strefler was asked whether it makes sense to assume our children and grandchildren will pay big bucks to suck CO2 out of the air when today's governments will not spend pennies to avoid emitting that pollution. Her reply: "If a country is not willing to do that, I don't see how they would be willing to" pay for CO2 removal.

Van Vuuren said his team is working to revise some assumptions, but he generally sees no technical reason CCS cannot work, or should be excluded from the possibilities his model considers. He said it is a policy decision whether countries wish to invest the dollars and effort to make it happen. The modelers' broader argument is that their job is to assemble a cookbook, not tell us how each recipe tastes. The models merely show that if CCS works and is used at an extraordinary scale, it is possible to hit climate targets.

For executives, that is the uncomfortable part. Net-zero pledges, carbon offsets and "climate pledge friendly" labels increasingly sit inside disclosure regimes, procurement standards and consumer marketing claims that regulators and litigants are starting to test. If the underlying assumption is a 92% capture rate that has never been demonstrated at scale, the gap between promise and performance becomes a balance-sheet and reputational exposure, not just a scientific debate. The modelers pointed to a mix of legitimate studies, unreviewed white papers, casual chats with industry insiders and the 2005 UN report, of which van Vuuren was an author. That is a thin evidentiary base for a multi-trillion-dollar capital allocation. Boards underwriting CCS-heavy transition plans, and the lenders and insurers behind them, now have a documented reason to stress-test the assumption rather than inherit it.

Executive ActionsLocked

This story's Key Insights and Take-aways are locked.

Create a free account to unlock Executive Actions for one credit.

Register to Unlock

Always free for Executives Club members. Join the Club

More in Science